Reactive power compensation for a Control Power Transformer (CPT) 3

[bacon4life]

What are typical reactive excitation losses in a 480V/120V single phase 500 VA Control Power Transformer for a motor control center? Would 60 VAr be reasonable?

Can I put a 120V capacitor in the control panel to provide power factor correction? The MCC already has power factor correction downstream of the motor contactor, but when the motor does not run the power factor appears to be very poor.

 

[LionelHutz]

How is the demand ratchet calculated?

How does your savings calculation work for all those months with 0kW demand and 0kVA PF adjusted demand?

0.1kVAr should be about 72kVArh a month, not 100kVArh a month.

 

[waross]

As I understood it, the 100 KVARHrs of correction was a value chosen for the comparison bill. It is close enough to the actual 72 KVARHrs to eliminate most of the PF issues, as is shown by the sample bill.
I apologize also. My thoughts mirror Skog's.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[bacon4life]

Oops, I put 100 in the wrong spot so the spreadsheet does show 100 kVArh rather 72 kVArh. I'll have to find out how reverse flowing VArs get metered to see whether overcorrection would create a billing issue.

The ratchet is a percentage of highest pf adjusted demand in the last 11 months, so reducing adjusted demand in month 7 (135 kVA to 63 kVA) would produce savings over the most recent 6 months shown.

For customers with normal usage patterns, the difference between using average vs peak PF would be much smaller. As with any kind of simple rule or law, there are edge cases like this that produce counterintuitive results.

Back to the original question though, is 12% reactive excitation loss in a CPT reasonable? I am used to substation transformers with much lower excitation current and don't have physical access to take direct measurement. If the reactive power isn't excitation losses, I need to figure out what else in a motor control center might consume VArs.

 

[davidbeach]

Just a thought, but do you know that it is actually a lagging PF that's getting you? Could it be that your parasitic capacitances are already over compensating the CPTs?

 

[waross]

The numbers with 0.10 VAR of correction look good. I'd use that.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

So how does the month 7 135 become 77 and 63 becomes 41?

Why don't the "corrected" months following month 7 all have the same ratchet and penalty kVA?

Why does the ratchet in month 17 change when the previous 11 months had the highest demand at month 7? By your description, the ratchet shouldn't drop until month 18 or 19 depending on how you count the 11 months previous (11 before billing month or including billing month).

Was the month with 47338kWh of demand billed over 39+ days? You've got significantly more kWh than a 50kW load would use in a month.

Are you going after the motor PF correction too? It appears that months of heavy motor/pump usage aren't running a high enough power factor either.

What is at the site? It's rare to see a pumping station where just a CPT is on-line without any load. Typically, there are some lights or PLC/SCADA or security system or some other stuff in most pumping stations.

You will have to try the capacitor. You'll first have to measure the CPT to figure out what size of capacitor to use.

davidbeach's suggestion of using DC controls is good too. Use a line powered 24VDC power supply which has power factor correction and change the coils of any relays or contactors to 24VDC. It'd probably draw less W and VA at idle.

Why does the demand kW vary on a pump? I would have expected it to be close to the same through those heavy usage months since the pump must have been running for long periods of time.

Is there a way to lower the demand kW? Everything else also hinges on that number, so if it can be consistently lower then it drops all the following numbers.

 

[waross]

Do you also have a lighting transformer? That may be demanding VARs as well.
Nevertheless, if the sample bill is valid, use the 0.10 Var correction to size your capacitor.
The capacitor may be on either the primary of secondary side of the transformer.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[bacon4life]

@Lionel-The newest reading is at the top of the sheet, so the months above the 135 kVA reading all have a ratchet of 81 kVA. There are several sites with similar billing issues, but so far the only site drawing I have shows just the CPT for each motor (15 & 75) with lighting provided by a separate utility service. The drawing is more than 25 years old, so further investigation is definitely in order. Most of the sites do have motor power factor capacitors already installed, though in this case they only got pf up to around 90%. During installation of a new small capacitor, I'll have them verify the fuses are still intact on all of the larger capacitors.

@David- Good point. I'll have to check on cable capacitance. Are the contactors always above ground so that the cables down the well are only energized while the pump is running? So far I have no idea whether these are submersible or surface mounted pumps.

 

[itsmoked]

I'd be looking at modest solar to monitor and control operations and a propane generator called automatically into service by the solar when needed. I bet you'd have a sub 2 year payback.

POCO be gone!

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

Nice "outside the box" thinking, Keith. I would need to see more details but that may work.
If the generator may be started directly connected to the pump, you would need only about 125% capacity compared to the 300% needed for DOL starting.
Looking at your billing, a 100 kW / 125 KVA generator would handle any of those months if the generator was started with the pump motor directly connected.
This brings the voltage and frequency up together similar to a VFD.
There are a few details to work out in the control and excitation scheme.
If there are a lot of sites you may consider one extra generator permanently mounted on a small truck that may be used as backup for any sites with problems.
Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LionelHutz]

I'm done with this. Apparently you don't want help because you refuse to answer how the demand ratchet is calculated. It isn't a hard question yet it GREATLY affects many months of billing.

 

[jghrist]

So how does the month 7 135 become 77 and 63 becomes 41?

Why don't the "corrected" months following month 7 all have the same ratchet and penalty kVA?

Why does the ratchet in month 17 change when the previous 11 months had the highest demand at month 7? By your description, the ratchet shouldn't drop until month 18 or 19 depending on how you count the 11 months previous (11 before billing month or including billing month).

Apparently you don't want help because you refuse to answer how the demand ratchet is calculated.

The bills are listed with the latest month on top. On the 16th row, the adjusted demand of 128 exceeds the previous 75 kVA ratchet. The ratchet after this becomes 60% of 128 (77). This is the ratchet until the adjusted demand becomes 135 in the 6th row and the ratchet becomes 60% of 135.

 

[bacon4life]

What is a typical cost for generators? I though the ballpark was around $1/watt, which would be quite a bit longer than a 2 year payback. Changing out the entire power supply scheme may be worth considering in the longer term capital budget cycle, but adding a small power factor capacitor to each site is likely something we could get implemented at all sites within a couple of months.

For periodic testing, a mobile generator something under consideration already. The cost of renting a generator plus the cost for a crew for the whole day to connect it, run the test and disconnect it would probably be more expensive than a single month's demand charge but less than the total ratcheted cost.

@Lionel- Sorry, the demand ratchet is 60% of the highest previous 11 months. Even without the demand ratchet, addressing this issue clearly has a short a payback period. I skipped the details of the billing at first because I had hoped to solicit technical expertise from the forum about any gotchas of providing power factor correction on small, unloaded transformers. The exact budgetary calculations are something the bean counters will check in detail for me before approving the project .

 

[LionelHutz]

Thanks for the info.

Without the demand ratchet you would hardly have an issue. I see about 3 or 4 months where a savings would be possible if it didn't exist.

A portable generator for testing won't be any use due to the demand ratchet.

If you have a lighting panel sourced from a separate meter then you could take the control power off that service instead. That would eliminate trying to figure out how to correct the transformers for the best possible results.